18331846. COLLECTIVE COMMUNICATION AS A MULTI-COMMODITY FLOW PROBLEM simplified abstract (Microsoft Technology Licensing, LLC)
Contents
COLLECTIVE COMMUNICATION AS A MULTI-COMMODITY FLOW PROBLEM
Organization Name
Microsoft Technology Licensing, LLC
Inventor(s)
Behnaz Arzani of Redmond WA (US)
Siva Kesava Reddy Kakarla of Bellevue WA (US)
Miguel Oom Temudo De Castro of Cambridge (GB)
Srikanth Kandula of Redmond WA (US)
Saeed Maleki of Seattle WA (US)
Luke Jonathon Marshall of Redmond WA (US)
COLLECTIVE COMMUNICATION AS A MULTI-COMMODITY FLOW PROBLEM - A simplified explanation of the abstract
This abstract first appeared for US patent application 18331846 titled 'COLLECTIVE COMMUNICATION AS A MULTI-COMMODITY FLOW PROBLEM
Simplified Explanation:
The patent application describes a method for efficiently scheduling the transfer of data among multiple processor nodes on a network. This method involves using a multi-commodity flow model to determine the optimal paths for data transfer while minimizing costs.
- The model takes input on the amount of data to be transferred to each processor node.
- It assigns paths connecting the processor nodes.
- It generates a schedule for data transfer along these paths, aiming to minimize costs by including store-and-forward and copy operations.
Key Features and Innovation:
- Utilizes a multi-commodity flow model to optimize data transfer scheduling.
- Minimizes costs by efficiently assigning paths and operations for data transfer.
- Incorporates store-and-forward and copy operations in the scheduling process.
Potential Applications:
- Data centers
- Cloud computing networks
- Distributed computing systems
Problems Solved:
- Inefficient data transfer scheduling
- High costs associated with data transfer
- Network congestion and bottlenecks
Benefits:
- Cost savings
- Improved network efficiency
- Reduced data transfer times
Commercial Applications:
The technology could be applied in various industries such as telecommunications, finance, and healthcare to optimize data transfer processes and reduce operational costs.
Prior Art:
Readers interested in prior art related to this technology could explore research papers on multi-commodity flow models and data transfer optimization algorithms.
Frequently Updated Research:
Researchers are constantly exploring new algorithms and models to further enhance the efficiency of data transfer scheduling in network systems.
Questions about Data Transfer Scheduling Technology:
1. How does the multi-commodity flow model differ from other optimization models used in data transfer scheduling? 2. What are the potential challenges in implementing this technology in real-world network systems?
Original Abstract Submitted
A method for scheduling a coordinated transfer of data among a plurality of processor nodes on a network comprises operating a multi-commodity flow model subject to a plurality of predetermined constraints. The model is configured to (a) receive as input a set of demands defining, for each of the plurality of processor nodes, an amount of data to be transferred to that processor node, (b) assign a plurality of paths linking the plurality of processor nodes, and (c) emit a schedule for transfer of the data along the plurality of paths so as to minimize a predetermined cost function, wherein the schedule comprises at least one store-and-forward operation and at least one copy operation.
